The innovation disclosed and claimed herein, in one aspect thereof, comprises systems and methods of managing regulatory questions. The systems and methods receive a question having words and phrases. The systems and methods identify keywords in the question using a knowledgebase. The systems and methods determine closely related questions based on the identification, the closely related questions having answers associated with each question of the closely related questions. The systems and methods perform machine learning on the answers of the determined closely related questions.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A method, comprising: one or more processors having one or more instructions comprising: receiving a question having words and phrases; identifying keywords in the question using a knowledgebase; determining a set of related questions based on the identification, the set of related questions are associated with a corresponding set of answers to each question of the set of related questions; performing machine learning on the question, the set of related questions, and answers of the determined set of related questions; and identifying a systemic risk based on the machine learning of the set of related questions received over a time period, wherein identifying the systemic risk comprises: determining keyword risk scores based on the identified keywords in the question and instances of the keywords in the set of related questions; identifying themes based on the keyword risk scores; and determining the systemic risk based on the identified themes and analysis of the identified themes across two or more lines of business of a financial institution.
2. The method of claim 1 , wherein the determining comprises: calculating a similarity score for each question in the set of related question; and selecting an answer associated with a highest scoring question as the answer to the received question.
3. The method of claim 2 , wherein the scoring comprises: converting the question into a vector of keywords or phrases.
4. The method of claim 3 , comprising: calculating a distance score between the vector of the question and vectors of the set of related questions stored in the knowledgebase.
5. The method of claim 3 , wherein the vector includes a count of each keyword or phrase in the question.
6. The method of claim 2 , comprising: approving the selected answer by a user for submission; and submitting the selected answer to a requesting authority.
7. The method of claim 6 , comprises: storing the submitted answers and question in the knowledgebase; and performing machine learning on the stored answers and questions in the knowledgebase to facilitate answering future questions.
8. The method of claim 1 , wherein the set of related questions and associated answers are stored in the knowledgebase.
9. The method of claim 1 , wherein the converting includes: parsing the words and phrases against the knowledgebase, the knowledgebase including synonyms and similar keywords or phrases.
10. The method of claim 1 , comprising: receiving an inquiry having a set of questions; and iteratively answering each question of the set of questions by determining the set of related questions and associated answers.
11. A system, comprising: one or more processors having instructions to control one or more components, comprising: a request component that receives a question having words and phrases; an identification component that identifies keywords in the question using a knowledgebase; a determination component that determines a set of related questions based on the identification, the set of related questions are associated with a corresponding set of answers to each question of the set of related questions; and a machine learning component that: performs machine learning on the question, the set of related questions, and answers of the determined set of related questions; and identifies a systemic risk based on the machine learning of the set of related questions received over a time period, wherein identifying the systemic risk comprises: determining keyword risk scores based on the identified keywords in the question and instances of the keywords in the set of related questions; identifying themes based on the keyword risk scores; and determining the systemic risk based on the identified themes and analysis of the identified themes across two or more lines of business of a financial institution.
12. The system of claim 11 , wherein the determination component comprises: a scoring component that calculates a similarity score for each question of the set of related questions; and a rank component that selects an answer associated with a highest scoring question as the answer to the received question.
13. The system of claim 12 , wherein the scoring comprises: a vector component that converts the question into a vector of keywords or phrases.
14. The system of claim 13 , comprising: a calculation component that calculates a distance score between the vector of the question and vectors of the set of related questions stored in the knowledgebase, wherein the vector includes a count of each keyword or phrase in the question.
15. The system of claim 12 , comprising: a graphical user interface to: receive approval of the selected answer for submission; and submit the selected answer to a requesting authority.
16. The system of claim 15 , wherein the knowledgebase stores the submitted answers for future questions.
17. The system of claim 16 , comprising: wherein the machine learning component analyzes the knowledgebase and submitted answers to improve selection of answers for future questions.
18. The system of claim 11 , comprising: wherein the identification component parses the words and phrases against the knowledgebase, the knowledgebase including synonyms and similar keywords or phrases.
19. A non-transitory computer readable medium having instructions to control one or more processors configured to: receive a question having words and phrases; identify keywords in the question using a knowledgebase; determine a set of related questions based on the identification, the set of related questions are associated with a corresponding set of answers to each question of the set of related questions; perform machine learning on the question, the set of related questions, and answers of the determined set of related questions; and identify a systemic risk based on the machine learning of the set of related questions received over a time period, wherein identifying the systemic risk comprises: determining keyword risk scores based on the identified keywords in the question and instances of the keywords in the set of related questions; identifying themes based on the keyword risk scores; and determining the systemic risk based on the identified themes and analysis of the identified themes across two or more lines of business of a financial institution.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
July 6, 2017
March 23, 2021
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.